Add TensorRT-LLM engine

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+engines/rank0.engine filter=lfs diff=lfs merge=lfs -text

engines/config.json

@@ -0,0 +1,93 @@
+{
+    "version": "0.9.0.dev2024022700",
+    "pretrained_config": {
+        "architecture": "GemmaForCausalLM",
+        "dtype": "bfloat16",
+        "logits_dtype": "float32",
+        "vocab_size": 256000,
+        "max_position_embeddings": 8192,
+        "hidden_size": 2048,
+        "num_hidden_layers": 18,
+        "num_attention_heads": 8,
+        "num_key_value_heads": 1,
+        "head_size": 256,
+        "hidden_act": "gelu",
+        "intermediate_size": 16384,
+        "norm_epsilon": 1e-06,
+        "position_embedding_type": "rope_gpt_neox",
+        "use_prompt_tuning": false,
+        "use_parallel_embedding": false,
+        "embedding_sharding_dim": 0,
+        "share_embedding_table": false,
+        "mapping": {
+            "world_size": 1,
+            "tp_size": 1,
+            "pp_size": 1
+        },
+        "kv_dtype": "bfloat16",
+        "max_lora_rank": 64,
+        "mlp_bias": false,
+        "attn_bias": false,
+        "rotary_base": 10000.0,
+        "rotary_scaling": null,
+        "enable_pos_shift": false,
+        "dense_context_fmha": false,
+        "use_fused_mlp": false,
+        "moe_num_experts": 0,
+        "moe_top_k": 0,
+        "moe_tp_mode": 2,
+        "moe_normalization_mode": 1,
+        "quantization": {
+            "quant_algo": null,
+            "kv_cache_quant_algo": null,
+            "group_size": null,
+            "has_zero_point": false,
+            "pre_quant_scale": false,
+            "exclude_modules": [],
+            "sq_use_plugin": false
+        }
+    },
+    "build_config": {
+        "max_input_len": 128,
+        "max_output_len": 8064,
+        "max_batch_size": 1,
+        "max_beam_width": 1,
+        "max_num_tokens": 128,
+        "max_prompt_embedding_table_size": 0,
+        "gather_context_logits": false,
+        "gather_generation_logits": false,
+        "strongly_typed": false,
+        "builder_opt": 3,
+        "profiling_verbosity": "layer_names_only",
+        "enable_debug_output": false,
+        "max_draft_len": 0,
+        "plugin_config": {
+            "bert_attention_plugin": null,
+            "gpt_attention_plugin": "bfloat16",
+            "gemm_plugin": "bfloat16",
+            "smooth_quant_gemm_plugin": null,
+            "identity_plugin": null,
+            "layernorm_quantization_plugin": null,
+            "rmsnorm_quantization_plugin": null,
+            "nccl_plugin": null,
+            "lookup_plugin": null,
+            "lora_plugin": null,
+            "weight_only_groupwise_quant_matmul_plugin": null,
+            "weight_only_quant_matmul_plugin": null,
+            "quantize_per_token_plugin": false,
+            "quantize_tensor_plugin": false,
+            "moe_plugin": null,
+            "context_fmha": true,
+            "context_fmha_fp32_acc": false,
+            "paged_kv_cache": true,
+            "remove_input_padding": true,
+            "use_custom_all_reduce": true,
+            "multi_block_mode": false,
+            "enable_xqa": true,
+            "attention_qk_half_accumulation": false,
+            "tokens_per_block": 128,
+            "use_paged_context_fmha": false,
+            "use_context_fmha_for_generation": false
+        }
+    }
+}

engines/rank0.engine

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa637ebf05e765053da8e2e02e14e8614d3fa64c81cceb074e256b92b5b16a70
+size 6063171228